In [1]:
%tensorflow_version 2.x
import tensorflow
tensorflow.__version__
Out[1]:
'2.3.0'
In [2]:
%matplotlib inline
import pandas as pd
import numpy as np
import tensorflow as tf
from google.colab import files
import seaborn as sns
from sklearn.model_selection import train_test_split
from tensorflow.keras import Sequential
from tensorflow.keras.layers import Dense
from sklearn import metrics
from sklearn.metrics import accuracy_score, confusion_matrix, precision_score, recall_score, f1_score, precision_recall_curve, auc
uploaded = files.upload()
uploaded = files.upload()
/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead. import pandas.util.testing as tm
Saving Labels.csv to Labels.csv
Saving images.npy to images.npy
Pre-Processing Image Data
In [3]:
import numpy as np
Data = pd.read_csv("Labels.csv")
Data.shape
Out[3]:
(4750, 1)
In [4]:
img_array = np.load("images.npy", allow_pickle=True)
In [5]:
img_array.shape
Out[5]:
(4750, 128, 128, 3)
In [6]:
from matplotlib import pyplot as plt
fig, axes = plt.subplots(10, 10, figsize=(100,100))
for i, ax in enumerate(axes.flat):
ax.imshow(img_array[i])
Visualizing of Images
In [7]:
X_data = np.array(img_array[:,:,0,0])
In [8]:
X_data.shape
Out[8]:
(4750, 128)
In [9]:
y_data = Data
In [10]:
y_data.shape
Out[10]:
(4750, 1)
In [11]:
X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size = 0.3, random_state = 7)
In [12]:
from sklearn import preprocessing
X_train = preprocessing.normalize(X_train)
In [13]:
print(X_train.shape)
print(X_test.shape)
print(y_train.shape)
print(y_test.shape)
(3325, 128) (1425, 128) (3325, 1) (1425, 1)
In [14]:
X_train = X_train.astype('float32')
X_test = X_test.astype('float32')
X_train /= 255
X_test /= 255
In [15]:
print("X_train shape:", X_train.shape)
print("Images in X_train:", X_train.shape[0])
print("Images in X_test:", X_test.shape[0])
print("Max value in X_train:", X_train.max())
print("Min value in X_train:", X_train.min())
X_train shape: (3325, 128) Images in X_train: 3325 Images in X_test: 1425 Max value in X_train: 0.0014285049 Min value in X_train: 0.0
In [16]:
import cv2
from matplotlib import pyplot as plt
img_array = np.load("images.npy", allow_pickle=True)
fig, axes = plt.subplots(10, 10, figsize=(100,100))
for i, ax in enumerate(axes.flat):
gaussian = cv2.GaussianBlur(img_array[i], (15, 15), 0)
ax.imshow(gaussian)
Data Compatibility
In [17]:
from sklearn.preprocessing import LabelBinarizer
enc = LabelBinarizer()
y_train = enc.fit_transform(y_train)
y_test = enc.fit_transform(y_test)
print("Shape of y_train:", y_train.shape)
print("One value of y_train:", y_train[0])
Shape of y_train: (3325, 12) One value of y_train: [0 0 0 1 0 0 0 0 0 0 0 0]
In [18]:
X_test, X_validation, y_test, y_validation = train_test_split(X_test, y_test, test_size = 0.5, random_state = 7)
In [19]:
validation_data = (X_validation, y_validation)
In [20]:
print(X_train.shape)
print(X_test.shape)
print(y_train.shape)
print(y_test.shape)
(3325, 128) (712, 128) (3325, 12) (712, 12)
In [21]:
print(X_train[1])
[0.00057647 0.00054154 0.0005328 0.0005328 0.00051533 0.00048913 0.00046293 0.00051533 0.0005328 0.00048913 0.00044546 0.00040178 0.00041052 0.00042799 0.00042799 0.00040178 0.00039305 0.00034938 0.00028824 0.00035811 0.00037558 0.00035811 0.00032317 0.00035811 0.00036685 0.00037558 0.00038432 0.00038432 0.00039305 0.00036685 0.00034938 0.00035811 0.00033191 0.00036685 0.00033191 0.00033191 0.00029697 0.0002795 0.00027077 0.00026203 0.0002795 0.00028824 0.00028824 0.00034938 0.00041052 0.00047166 0.00047166 0.00038432 0.00019216 0.00020089 0.00030571 0.00032317 0.00030571 0.00032317 0.00033191 0.00032317 0.00040178 0.00033191 0.00028824 0.00032317 0.00032317 0.00032317 0.00031444 0.00031444 0.00031444 0.00030571 0.00029697 0.00030571 0.00029697 0.00031444 0.00043672 0.00048039 0.00037558 0.00033191 0.00031444 0.00030571 0.00028824 0.00028824 0.0002533 0.00023583 0.00026203 0.0002795 0.00027077 0.0002271 0.0002271 0.0002271 0.0002271 0.00023583 0.00023583 0.0002533 0.0002533 0.00024456 0.0002271 0.00023583 0.0002795 0.00030571 0.00035811 0.00039305 0.00035811 0.0002795 0.00034938 0.00029697 0.00027077 0.00034938 0.00031444 0.00032317 0.00021836 0.00020963 0.00026203 0.00032317 0.00031444 0.00028824 0.00033191 0.00035811 0.00034938 0.00029697 0.0002533 0.00026203 0.00026203 0.00034064 0.00038432 0.00038432 0.00039305 0.00041925 0.00033191 0.00028824 0.00034064 0.00032317]
In [22]:
X_train = X_train.reshape((X_train.shape[0], 128)).astype('float32')
X_test = X_test.reshape(X_test.shape[0], 128).astype('float32')
print(X_train.shape)
print(X_test.shape)
(3325, 128) (712, 128)
In [23]:
X_train = np.expand_dims(X_train, axis = 2)
print(X_train.shape)
print(X_test.shape)
(3325, 128, 1) (712, 128)
Building CNN
In [24]:
from tensorflow.keras import datasets, models, layers, optimizers
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from tensorflow.keras.callbacks import ModelCheckpoint, EarlyStopping
from google.colab.patches import cv2_imshow
In [25]:
# Set the CNN model
batch_size = None
model = models.Sequential()
model.add(layers.Conv2D(32, (5, 5), padding='same', activation="relu", input_shape=(128,128,1),data_format='channels_first'))
model.add(layers.BatchNormalization())
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Dropout(0.2))
model.add(layers.Conv2D(64, (5, 5), padding='same', activation="relu"))
model.add(layers.BatchNormalization())
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Dropout(0.3))
model.add(layers.Conv2D(64, (3, 3), padding='same', activation="relu"))
model.add(layers.BatchNormalization())
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Dropout(0.4))
model.add(layers.Conv2D(64, (3, 3), padding='same', activation="relu"))
model.add(layers.BatchNormalization())
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Dropout(0.5))
model.add(layers.GlobalMaxPooling2D())
model.add(layers.Dense(256, activation="relu"))
model.add(layers.Dropout(0.5))
model.add(layers.Dense(10, activation="softmax"))
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= conv2d (Conv2D) (None, 32, 128, 1) 102432 _________________________________________________________________ batch_normalization (BatchNo (None, 32, 128, 1) 4 _________________________________________________________________ max_pooling2d (MaxPooling2D) (None, 16, 64, 1) 0 _________________________________________________________________ dropout (Dropout) (None, 16, 64, 1) 0 _________________________________________________________________ conv2d_1 (Conv2D) (None, 16, 64, 64) 1664 _________________________________________________________________ batch_normalization_1 (Batch (None, 16, 64, 64) 256 _________________________________________________________________ max_pooling2d_1 (MaxPooling2 (None, 8, 32, 64) 0 _________________________________________________________________ dropout_1 (Dropout) (None, 8, 32, 64) 0 _________________________________________________________________ conv2d_2 (Conv2D) (None, 8, 32, 64) 36928 _________________________________________________________________ batch_normalization_2 (Batch (None, 8, 32, 64) 256 _________________________________________________________________ max_pooling2d_2 (MaxPooling2 (None, 4, 16, 64) 0 _________________________________________________________________ dropout_2 (Dropout) (None, 4, 16, 64) 0 _________________________________________________________________ conv2d_3 (Conv2D) (None, 4, 16, 64) 36928 _________________________________________________________________ batch_normalization_3 (Batch (None, 4, 16, 64) 256 _________________________________________________________________ max_pooling2d_3 (MaxPooling2 (None, 2, 8, 64) 0 _________________________________________________________________ dropout_3 (Dropout) (None, 2, 8, 64) 0 _________________________________________________________________ global_max_pooling2d (Global (None, 64) 0 _________________________________________________________________ dense (Dense) (None, 256) 16640 _________________________________________________________________ dropout_4 (Dropout) (None, 256) 0 _________________________________________________________________ dense_1 (Dense) (None, 10) 2570 ================================================================= Total params: 197,934 Trainable params: 197,548 Non-trainable params: 386 _________________________________________________________________
In [26]:
opt = optimizers.Adam(learning_rate=0.001, beta_1=0.9, beta_2=0.999, epsilon=1e-08)
In [27]:
model.compile(loss='categorical_crossentropy',
optimizer=opt,
metrics=['accuracy'])
In [28]:
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= conv2d (Conv2D) (None, 32, 128, 1) 102432 _________________________________________________________________ batch_normalization (BatchNo (None, 32, 128, 1) 4 _________________________________________________________________ max_pooling2d (MaxPooling2D) (None, 16, 64, 1) 0 _________________________________________________________________ dropout (Dropout) (None, 16, 64, 1) 0 _________________________________________________________________ conv2d_1 (Conv2D) (None, 16, 64, 64) 1664 _________________________________________________________________ batch_normalization_1 (Batch (None, 16, 64, 64) 256 _________________________________________________________________ max_pooling2d_1 (MaxPooling2 (None, 8, 32, 64) 0 _________________________________________________________________ dropout_1 (Dropout) (None, 8, 32, 64) 0 _________________________________________________________________ conv2d_2 (Conv2D) (None, 8, 32, 64) 36928 _________________________________________________________________ batch_normalization_2 (Batch (None, 8, 32, 64) 256 _________________________________________________________________ max_pooling2d_2 (MaxPooling2 (None, 4, 16, 64) 0 _________________________________________________________________ dropout_2 (Dropout) (None, 4, 16, 64) 0 _________________________________________________________________ conv2d_3 (Conv2D) (None, 4, 16, 64) 36928 _________________________________________________________________ batch_normalization_3 (Batch (None, 4, 16, 64) 256 _________________________________________________________________ max_pooling2d_3 (MaxPooling2 (None, 2, 8, 64) 0 _________________________________________________________________ dropout_3 (Dropout) (None, 2, 8, 64) 0 _________________________________________________________________ global_max_pooling2d (Global (None, 64) 0 _________________________________________________________________ dense (Dense) (None, 256) 16640 _________________________________________________________________ dropout_4 (Dropout) (None, 256) 0 _________________________________________________________________ dense_1 (Dense) (None, 10) 2570 ================================================================= Total params: 197,934 Trainable params: 197,548 Non-trainable params: 386 _________________________________________________________________
Evaluate the model.
Hi, last night, my model.fit function stopped working all of a sudden. I am not able to make even the mentor session examples work. In order for my code to compile, I am using a really simple model. This is my original code: model1.fit( x = X_train, y=y_train, batch_size=128, epochs=10, validation_split = 0.5). It worked up until some point, and then it is not. I tried classroom examples too and am getting errors. I can't make this code work with the simplified model: scores = model2.evaluate(x, y, verbose=1) print('Test loss:', scores[0]) print('Test accuracy:', scores[1]).
In [29]:
model2 = Sequential()
model2.add(Dense(1, input_shape=(1,)))
model2.compile(loss='mse', optimizer='rmsprop')
# The fit() method - trains the model
x = np.random.uniform(0.0, 1.0, (200))
y = 0.3 + 0.6*x + np.random.normal(0.0, 0.05,(200))
model2.fit(x, y, epochs=1000, batch_size=100)
Epoch 1/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.7606 Epoch 2/1000 2/2 [==============================] - 0s 998us/step - loss: 0.7483 Epoch 3/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.7396 Epoch 4/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.7323 Epoch 5/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.7257 Epoch 6/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.7195 Epoch 7/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.7137 Epoch 8/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.7081 Epoch 9/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.7027 Epoch 10/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6975 Epoch 11/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6923 Epoch 12/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6873 Epoch 13/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6823 Epoch 14/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6773 Epoch 15/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6725 Epoch 16/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6676 Epoch 17/1000 2/2 [==============================] - 0s 989us/step - loss: 0.6628 Epoch 18/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6581 Epoch 19/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6533 Epoch 20/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6486 Epoch 21/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6440 Epoch 22/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6393 Epoch 23/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6347 Epoch 24/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6301 Epoch 25/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6255 Epoch 26/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6210 Epoch 27/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6164 Epoch 28/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6119 Epoch 29/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.6074 Epoch 30/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.6029 Epoch 31/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5985 Epoch 32/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.5940 Epoch 33/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5896 Epoch 34/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5852 Epoch 35/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5808 Epoch 36/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5764 Epoch 37/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5720 Epoch 38/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.5677 Epoch 39/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5634 Epoch 40/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5591 Epoch 41/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5548 Epoch 42/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5505 Epoch 43/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.5463 Epoch 44/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5421 Epoch 45/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5378 Epoch 46/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5336 Epoch 47/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5295 Epoch 48/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5253 Epoch 49/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5212 Epoch 50/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5170 Epoch 51/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5129 Epoch 52/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5088 Epoch 53/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5048 Epoch 54/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.5007 Epoch 55/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4967 Epoch 56/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4927 Epoch 57/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4887 Epoch 58/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4847 Epoch 59/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4807 Epoch 60/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4768 Epoch 61/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4728 Epoch 62/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4689 Epoch 63/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4650 Epoch 64/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4611 Epoch 65/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4573 Epoch 66/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4534 Epoch 67/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4496 Epoch 68/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4458 Epoch 69/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4420 Epoch 70/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4382 Epoch 71/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4345 Epoch 72/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4307 Epoch 73/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4270 Epoch 74/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4233 Epoch 75/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4196 Epoch 76/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4160 Epoch 77/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4123 Epoch 78/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4087 Epoch 79/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.4051 Epoch 80/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.4015 Epoch 81/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3979 Epoch 82/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3943 Epoch 83/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3908 Epoch 84/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3873 Epoch 85/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3838 Epoch 86/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3803 Epoch 87/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3768 Epoch 88/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3734 Epoch 89/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3699 Epoch 90/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3665 Epoch 91/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3631 Epoch 92/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3597 Epoch 93/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3564 Epoch 94/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3530 Epoch 95/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3497 Epoch 96/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3464 Epoch 97/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3431 Epoch 98/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3398 Epoch 99/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3365 Epoch 100/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.3333 Epoch 101/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3300 Epoch 102/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3268 Epoch 103/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3236 Epoch 104/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3204 Epoch 105/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3173 Epoch 106/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3141 Epoch 107/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3110 Epoch 108/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3079 Epoch 109/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3048 Epoch 110/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.3017 Epoch 111/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2987 Epoch 112/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2956 Epoch 113/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2926 Epoch 114/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2896 Epoch 115/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2866 Epoch 116/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2837 Epoch 117/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2807 Epoch 118/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2778 Epoch 119/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2749 Epoch 120/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2720 Epoch 121/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2691 Epoch 122/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2662 Epoch 123/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2634 Epoch 124/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2605 Epoch 125/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2577 Epoch 126/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2549 Epoch 127/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2522 Epoch 128/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2494 Epoch 129/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2467 Epoch 130/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2440 Epoch 131/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2412 Epoch 132/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2386 Epoch 133/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2359 Epoch 134/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2332 Epoch 135/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2306 Epoch 136/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2280 Epoch 137/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2254 Epoch 138/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2228 Epoch 139/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2202 Epoch 140/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2177 Epoch 141/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2151 Epoch 142/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2126 Epoch 143/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2101 Epoch 144/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2076 Epoch 145/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.2052 Epoch 146/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2027 Epoch 147/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.2003 Epoch 148/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1979 Epoch 149/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1955 Epoch 150/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.1931 Epoch 151/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1908 Epoch 152/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1884 Epoch 153/1000 2/2 [==============================] - 0s 958us/step - loss: 0.1861 Epoch 154/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1838 Epoch 155/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1815 Epoch 156/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1792 Epoch 157/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1770 Epoch 158/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1747 Epoch 159/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1725 Epoch 160/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1703 Epoch 161/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1681 Epoch 162/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1659 Epoch 163/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1638 Epoch 164/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1616 Epoch 165/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.1595 Epoch 166/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.1574 Epoch 167/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.1553 Epoch 168/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1533 Epoch 169/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1512 Epoch 170/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1492 Epoch 171/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1472 Epoch 172/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1452 Epoch 173/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1432 Epoch 174/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1412 Epoch 175/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1393 Epoch 176/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1373 Epoch 177/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.1354 Epoch 178/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.1336 Epoch 179/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1317 Epoch 180/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.1298 Epoch 181/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1280 Epoch 182/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.1262 Epoch 183/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1244 Epoch 184/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1226 Epoch 185/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1208 Epoch 186/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1190 Epoch 187/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1173 Epoch 188/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1156 Epoch 189/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1139 Epoch 190/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1122 Epoch 191/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1105 Epoch 192/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1088 Epoch 193/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1072 Epoch 194/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1056 Epoch 195/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1040 Epoch 196/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1024 Epoch 197/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.1009 Epoch 198/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0993 Epoch 199/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0978 Epoch 200/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0963 Epoch 201/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0948 Epoch 202/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0933 Epoch 203/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0918 Epoch 204/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0904 Epoch 205/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0889 Epoch 206/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0875 Epoch 207/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0861 Epoch 208/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0847 Epoch 209/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0834 Epoch 210/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0820 Epoch 211/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0807 Epoch 212/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0794 Epoch 213/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0781 Epoch 214/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0768 Epoch 215/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0755 Epoch 216/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0743 Epoch 217/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0731 Epoch 218/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0718 Epoch 219/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0706 Epoch 220/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0695 Epoch 221/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0683 Epoch 222/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0672 Epoch 223/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0660 Epoch 224/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0649 Epoch 225/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0638 Epoch 226/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0627 Epoch 227/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0617 Epoch 228/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0606 Epoch 229/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0596 Epoch 230/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0586 Epoch 231/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0576 Epoch 232/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0566 Epoch 233/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0556 Epoch 234/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0547 Epoch 235/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0538 Epoch 236/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0528 Epoch 237/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0520 Epoch 238/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0511 Epoch 239/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0502 Epoch 240/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0494 Epoch 241/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0485 Epoch 242/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0477 Epoch 243/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0469 Epoch 244/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0461 Epoch 245/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0454 Epoch 246/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0446 Epoch 247/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0439 Epoch 248/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0432 Epoch 249/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0425 Epoch 250/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0418 Epoch 251/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0411 Epoch 252/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0404 Epoch 253/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0398 Epoch 254/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0392 Epoch 255/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0386 Epoch 256/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0380 Epoch 257/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0374 Epoch 258/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0368 Epoch 259/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0363 Epoch 260/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0357 Epoch 261/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0352 Epoch 262/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0347 Epoch 263/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0342 Epoch 264/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0338 Epoch 265/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0333 Epoch 266/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0329 Epoch 267/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0324 Epoch 268/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0320 Epoch 269/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0316 Epoch 270/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0312 Epoch 271/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0309 Epoch 272/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0305 Epoch 273/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0301 Epoch 274/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0298 Epoch 275/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0295 Epoch 276/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0292 Epoch 277/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0289 Epoch 278/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0286 Epoch 279/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0284 Epoch 280/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0281 Epoch 281/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0279 Epoch 282/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0276 Epoch 283/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0274 Epoch 284/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0272 Epoch 285/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0270 Epoch 286/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0268 Epoch 287/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0266 Epoch 288/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0264 Epoch 289/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0262 Epoch 290/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0260 Epoch 291/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0259 Epoch 292/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0257 Epoch 293/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0255 Epoch 294/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0254 Epoch 295/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0252 Epoch 296/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0251 Epoch 297/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0249 Epoch 298/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0247 Epoch 299/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0246 Epoch 300/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0245 Epoch 301/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0243 Epoch 302/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0241 Epoch 303/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0239 Epoch 304/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0238 Epoch 305/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0236 Epoch 306/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0235 Epoch 307/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0233 Epoch 308/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0231 Epoch 309/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0230 Epoch 310/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0228 Epoch 311/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0226 Epoch 312/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0225 Epoch 313/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0223 Epoch 314/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0221 Epoch 315/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0220 Epoch 316/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0218 Epoch 317/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0217 Epoch 318/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0215 Epoch 319/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0214 Epoch 320/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0212 Epoch 321/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0210 Epoch 322/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0209 Epoch 323/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0207 Epoch 324/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0206 Epoch 325/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0205 Epoch 326/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0203 Epoch 327/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0201 Epoch 328/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0200 Epoch 329/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0198 Epoch 330/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0197 Epoch 331/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0195 Epoch 332/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0194 Epoch 333/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0192 Epoch 334/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0191 Epoch 335/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0189 Epoch 336/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0188 Epoch 337/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0186 Epoch 338/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0185 Epoch 339/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0184 Epoch 340/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0182 Epoch 341/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0181 Epoch 342/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0179 Epoch 343/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0178 Epoch 344/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0176 Epoch 345/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0175 Epoch 346/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0174 Epoch 347/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0172 Epoch 348/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0171 Epoch 349/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0169 Epoch 350/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0168 Epoch 351/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0167 Epoch 352/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0165 Epoch 353/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0164 Epoch 354/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0163 Epoch 355/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0161 Epoch 356/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0160 Epoch 357/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0159 Epoch 358/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0157 Epoch 359/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0156 Epoch 360/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0155 Epoch 361/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0153 Epoch 362/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0152 Epoch 363/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0151 Epoch 364/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0149 Epoch 365/1000 2/2 [==============================] - 0s 10ms/step - loss: 0.0148 Epoch 366/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0147 Epoch 367/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0146 Epoch 368/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0144 Epoch 369/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0143 Epoch 370/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0142 Epoch 371/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0141 Epoch 372/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0140 Epoch 373/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0138 Epoch 374/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0137 Epoch 375/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0136 Epoch 376/1000 2/2 [==============================] - 0s 4ms/step - loss: 0.0135 Epoch 377/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0133 Epoch 378/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0132 Epoch 379/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0131 Epoch 380/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0130 Epoch 381/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0129 Epoch 382/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0128 Epoch 383/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0126 Epoch 384/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0125 Epoch 385/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0124 Epoch 386/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0123 Epoch 387/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0122 Epoch 388/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0121 Epoch 389/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0120 Epoch 390/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0119 Epoch 391/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0117 Epoch 392/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0116 Epoch 393/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0115 Epoch 394/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0114 Epoch 395/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0113 Epoch 396/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0112 Epoch 397/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0111 Epoch 398/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0110 Epoch 399/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0109 Epoch 400/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0108 Epoch 401/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0107 Epoch 402/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0106 Epoch 403/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0105 Epoch 404/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0104 Epoch 405/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0103 Epoch 406/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0102 Epoch 407/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0101 Epoch 408/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0100 Epoch 409/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0099 Epoch 410/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0098 Epoch 411/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0097 Epoch 412/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0096 Epoch 413/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0095 Epoch 414/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0094 Epoch 415/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0093 Epoch 416/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0092 Epoch 417/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0091 Epoch 418/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0090 Epoch 419/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0089 Epoch 420/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0088 Epoch 421/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0087 Epoch 422/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0087 Epoch 423/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0086 Epoch 424/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0085 Epoch 425/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0084 Epoch 426/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0083 Epoch 427/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0082 Epoch 428/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0081 Epoch 429/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0080 Epoch 430/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0079 Epoch 431/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0079 Epoch 432/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0078 Epoch 433/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0077 Epoch 434/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0076 Epoch 435/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0075 Epoch 436/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0075 Epoch 437/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0074 Epoch 438/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0073 Epoch 439/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0072 Epoch 440/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0071 Epoch 441/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0071 Epoch 442/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0070 Epoch 443/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0069 Epoch 444/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0068 Epoch 445/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0068 Epoch 446/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0067 Epoch 447/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0066 Epoch 448/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0066 Epoch 449/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0065 Epoch 450/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0064 Epoch 451/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0063 Epoch 452/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0063 Epoch 453/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0062 Epoch 454/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0061 Epoch 455/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0061 Epoch 456/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0060 Epoch 457/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0059 Epoch 458/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0059 Epoch 459/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0058 Epoch 460/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0058 Epoch 461/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0057 Epoch 462/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0056 Epoch 463/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0056 Epoch 464/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0055 Epoch 465/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0055 Epoch 466/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0054 Epoch 467/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0053 Epoch 468/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0053 Epoch 469/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0052 Epoch 470/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0051 Epoch 471/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0051 Epoch 472/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0050 Epoch 473/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0050 Epoch 474/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0049 Epoch 475/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0049 Epoch 476/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0048 Epoch 477/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0048 Epoch 478/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0047 Epoch 479/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0047 Epoch 480/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0046 Epoch 481/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0045 Epoch 482/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0045 Epoch 483/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0044 Epoch 484/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0044 Epoch 485/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0043 Epoch 486/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0043 Epoch 487/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0043 Epoch 488/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0042 Epoch 489/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0042 Epoch 490/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0041 Epoch 491/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0041 Epoch 492/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0040 Epoch 493/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0040 Epoch 494/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0039 Epoch 495/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0039 Epoch 496/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0039 Epoch 497/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0038 Epoch 498/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0038 Epoch 499/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0037 Epoch 500/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0037 Epoch 501/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0036 Epoch 502/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0036 Epoch 503/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0036 Epoch 504/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0035 Epoch 505/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0035 Epoch 506/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0035 Epoch 507/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0034 Epoch 508/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0034 Epoch 509/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0034 Epoch 510/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0033 Epoch 511/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0033 Epoch 512/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0033 Epoch 513/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0032 Epoch 514/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0032 Epoch 515/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0032 Epoch 516/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0032 Epoch 517/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0031 Epoch 518/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0031 Epoch 519/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0031 Epoch 520/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0030 Epoch 521/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0030 Epoch 522/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0030 Epoch 523/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0030 Epoch 524/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0029 Epoch 525/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0029 Epoch 526/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0029 Epoch 527/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0029 Epoch 528/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0028 Epoch 529/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0028 Epoch 530/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0028 Epoch 531/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0028 Epoch 532/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0028 Epoch 533/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0027 Epoch 534/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0027 Epoch 535/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0027 Epoch 536/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0027 Epoch 537/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0027 Epoch 538/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0027 Epoch 539/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0026 Epoch 540/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0026 Epoch 541/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0026 Epoch 542/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0026 Epoch 543/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0026 Epoch 544/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0026 Epoch 545/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0026 Epoch 546/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0026 Epoch 547/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0026 Epoch 548/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 549/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 550/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 551/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0025 Epoch 552/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 553/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 554/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 555/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 556/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 557/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 558/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 559/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 560/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0025 Epoch 561/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 562/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 563/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 564/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 565/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 566/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 567/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 568/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 569/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 570/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 571/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 572/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 573/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 574/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 575/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 576/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 577/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 578/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 579/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 580/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 581/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 582/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 583/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 584/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 585/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 586/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 587/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 588/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 589/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 590/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 591/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 592/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 593/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 594/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 595/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 596/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 597/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 598/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 599/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 600/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 601/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 602/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 603/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 604/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 605/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 606/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 607/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 608/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 609/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 610/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 611/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 612/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 613/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 614/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 615/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 616/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 617/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 618/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 619/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 620/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 621/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 622/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 623/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 624/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 625/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 626/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 627/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 628/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 629/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 630/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 631/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 632/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 633/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 634/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 635/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 636/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 637/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 638/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 639/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 640/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 641/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 642/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 643/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 644/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 645/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 646/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 647/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 648/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 649/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 650/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 651/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 652/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 653/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 654/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 655/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 656/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 657/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 658/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 659/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 660/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 661/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 662/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 663/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 664/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 665/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 666/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 667/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 668/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 669/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 670/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 671/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 672/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 673/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 674/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 675/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 676/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 677/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 678/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 679/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 680/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 681/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 682/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 683/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 684/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 685/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 686/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 687/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 688/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 689/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 690/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 691/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 692/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 693/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 694/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 695/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 696/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 697/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 698/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 699/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 700/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 701/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 702/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 703/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 704/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 705/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 706/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 707/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 708/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 709/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 710/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 711/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 712/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 713/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 714/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 715/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 716/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 717/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 718/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 719/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 720/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 721/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 722/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 723/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 724/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 725/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 726/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 727/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 728/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 729/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 730/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 731/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 732/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 733/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 734/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 735/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 736/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 737/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 738/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 739/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 740/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 741/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 742/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 743/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 744/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 745/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 746/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 747/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 748/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 749/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 750/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 751/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 752/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 753/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 754/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 755/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 756/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 757/1000 2/2 [==============================] - 0s 4ms/step - loss: 0.0024 Epoch 758/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 759/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 760/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 761/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 762/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 763/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 764/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 765/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 766/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 767/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 768/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 769/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 770/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 771/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 772/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 773/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 774/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 775/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 776/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 777/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 778/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 779/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 780/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 781/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 782/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 783/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 784/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 785/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 786/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 787/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 788/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 789/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 790/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 791/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 792/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 793/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 794/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 795/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 796/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 797/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 798/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 799/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 800/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 801/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 802/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 803/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 804/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 805/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 806/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 807/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 808/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 809/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 810/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 811/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 812/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 813/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 814/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 815/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 816/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 817/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 818/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 819/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 820/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 821/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 822/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 823/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 824/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 825/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 826/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 827/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 828/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 829/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 830/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 831/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 832/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 833/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 834/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 835/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 836/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 837/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 838/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 839/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 840/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 841/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 842/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 843/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 844/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 845/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 846/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 847/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 848/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 849/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 850/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 851/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 852/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 853/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 854/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 855/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 856/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 857/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 858/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 859/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 860/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 861/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 862/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 863/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 864/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 865/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 866/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 867/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 868/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 869/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 870/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 871/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 872/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 873/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 874/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 875/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 876/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 877/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 878/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 879/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 880/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 881/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 882/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 883/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 884/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 885/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 886/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 887/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 888/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 889/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 890/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 891/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 892/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 893/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 894/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 895/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 896/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 897/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 898/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 899/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 900/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 901/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 902/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 903/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 904/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 905/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 906/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 907/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 908/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 909/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 910/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 911/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 912/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 913/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 914/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 915/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 916/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 917/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 918/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 919/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 920/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 921/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 922/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 923/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 924/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 925/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 926/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 927/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 928/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 929/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 930/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 931/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 932/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 933/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 934/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 935/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 936/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 937/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 938/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 939/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 940/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 941/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 942/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 943/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 944/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 945/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 946/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 947/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 948/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 949/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 950/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 951/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 952/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 953/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 954/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 955/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 956/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 957/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 958/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 959/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 960/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 961/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 962/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 963/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 964/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 965/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 966/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 967/1000 2/2 [==============================] - 0s 7ms/step - loss: 0.0024 Epoch 968/1000 2/2 [==============================] - 0s 3ms/step - loss: 0.0024 Epoch 969/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 970/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 971/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 972/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 973/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 974/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 975/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 976/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 977/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 978/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 979/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 980/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 981/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 982/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 983/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 984/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 985/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 986/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 987/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 988/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 989/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 990/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 991/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 992/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 993/1000 2/2 [==============================] - 0s 1ms/step - loss: 0.0024 Epoch 994/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 995/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 996/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 997/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 998/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 999/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024 Epoch 1000/1000 2/2 [==============================] - 0s 2ms/step - loss: 0.0024
Out[29]:
<tensorflow.python.keras.callbacks.History at 0x7fe8faf85f98>
In [30]:
results = model2.evaluate(x, y)
Y_pred_cls = model2.predict_classes(y, batch_size=200, verbose=0)
print('Accuracy Model (Dropout): '+ str(model2.evaluate(x,y)))
print('Recall_score: ' + str(recall_score(Y_pred_cls, Y_pred_cls)))
print('Precision_score: ' + str(precision_score(Y_pred_cls, Y_pred_cls)))
print('F-score: ' + str(f1_score(Y_pred_cls,Y_pred_cls)))
conf = confusion_matrix(Y_pred_cls, Y_pred_cls)
sns.heatmap(conf.T, square=True, annot=True, cbar=False, cmap=plt.cm.Blues)
plt.xlabel('Predicted Values')
plt.ylabel('True Values');
plt.show();
7/7 [==============================] - 0s 1ms/step - loss: 0.0024
WARNING:tensorflow:From <ipython-input-30-c2411fccecb2>:2: Sequential.predict_classes (from tensorflow.python.keras.engine.sequential) is deprecated and will be removed after 2021-01-01.
Instructions for updating:
Please use instead:* `np.argmax(model.predict(x), axis=-1)`, if your model does multi-class classification (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype("int32")`, if your model does binary classification (e.g. if it uses a `sigmoid` last-layer activation).
7/7 [==============================] - 0s 1ms/step - loss: 0.0024
Accuracy Model (Dropout): 0.002388233318924904
Recall_score: 1.0
Precision_score: 1.0
F-score: 1.0
In [31]:
from keras.utils.vis_utils import plot_model
plot_model(model2, to_file='model_plot.png', show_shapes=True, show_layer_names=True)
Out[31]:
In [32]:
X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.9, random_state=0)
y_pred = model2.predict(X_test)
import numpy as np
from matplotlib import pyplot as plt
data = np.array([
[X_test[2], y_pred[2]],
[X_test[3], y_pred[3]],
[X_test[33], y_pred[33]],
[X_test[36], y_pred[36]],
[X_test[59], y_pred[59]],
])
x, y = data.T
plt.scatter(x,y)
plt.show()
